R-alpha: libf2c as a shared library

[Ross Ihaka]

Ok, I did the following experiment

	dyn.load("/usr/local/lib/libf2c.so.2.0")
	dyn.load("try.so")
	.Fortran("try", 10.5)

where try.so contains the following compiled subroutine

	subroutine try(x)
        double precision x
        write(*,*) x
        return
        end

It worked just fine (at least here on my FreeBSD machine at home).
So it seems that it will be easy to make all of f2c available,
including the all the IO stuff.

It should be easy to package the libf2c sources as an R library.
I'm hesitant to put it into the core R distribution, because it
adds a lot of source, although this would ensure that it was always
available.

So, should it go in the R core, or to CRAN?
	Ross
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-